Labdane and Clerodane Diterpenoids from - ACS Publications

Oct 7, 2015 - Directorate Research Development, University of the Free State, ...... C.; Hering, S.; Hamburger, M. Phytochemistry 2013, 96, 318−329...
1 downloads 0 Views 3MB Size
Article pubs.acs.org/jnp

Labdane and Clerodane Diterpenoids from Colophospermum mopane Kun Du,†,∞ Maria De Mieri,‡,∞ Markus Neuburger,§ Pieter C. Zietsman,⊥ Andrew Marston,†,# Sandy F. van Vuuren,∥ Daneel Ferreira,▽ Matthias Hamburger,*,‡ and Jan H. van der Westhuizen*,○ †

Department of Chemistry and ○Directorate Research Development, University of the Free State, Bloemfontein 9301, South Africa Division of Pharmaceutical Biology and §Division of Inorganic Chemistry, Department of Chemistry, University of Basel, 4056 Basel, Switzerland ⊥ National Museum, P.O. Box 266, Bloemfontein 9300, South Africa ∥ Department of Pharmacy and Pharmacology, Faculty of Health Sciences, University of the Witwatersrand, Parktown 2193, South Africa ▽ Department of Biomolecular Sciences, Division of Pharmacognosy, and Research Institute of Pharmaceutical Sciences, School of Pharmacy, University of Mississippi, University, Mississippi 38677-1848, United States ‡

S Supporting Information *

ABSTRACT: Five labdane (1−5), an isolabdane (6), and five clerodane diterpenoids (7−11), were isolated from seeds, husks, and leaves of Colophospermum mopane. Compounds 1−3 and 6−9 are new, and their structures were elucidated by means of physical data analysis (1D and 2D NMR, HRESIMS). The absolute configurations of 1, 7, and 10 were determined by single-crystal X-ray diffraction with Cu Kα radiation. For compounds 2 and 6, the absolute configurations were established by the modified Mosher’s method and corroborated by comparison of experimental and calculated electronic circular dichroism spectra of their 3-pbromobenzoate derivatives. The crude extracts and compounds were evaluated for antimicrobial activity. The leaf extract was the most active against Staphylococcus aureus (125 μg/mL). Compound 11 showed the best inhibitory activity, with minimum inhibitory values of 15.6 μg/mL against Escherichia coli and Staphylococcus aureus and 31.3 μg/mL against Enterococcus faecalis. and leaves.6−8 As part of an ongoing search for novel bioactive compounds from medicinal plants in southern Africa,9−11 we focused on C. mopane, one of the most readily recognized trees because of its distinctive bifoliolate leaves and widespread distribution. Fractionation of the CH2Cl2 extracts of the seeds, husks, and leaves, respectively, afforded three new (1−3) and two known (4 and 5) labdane diterpenoids, a new isolabdane (6), and three new (7−9) and two known (10 and 11) clerodane diterpenoids. Herein are reported the isolation and structure elucidation of these diterpenoids and their antibacterial activity.

Colophospermum mopane Kirk ex J. Leonard (previously Copaifera mopane Kirk ex Benth), belonging to a monotypic genus in the Caesalpinioideae (Leguminosae subfamily),1 is a dominant tree in the dry regions of southern Africa. It is one of the most distinctive vegetation groups, often forming pure stands. These have given rise to the now accepted term “mopane woodland” or “Colophospermum woodland”, which has an atmosphere entirely of its own.2 The leaves and pods of C. mopane provide an important food source for many animals, while the roasted caterpillars of the mopane moth, commonly known as “mopane worms”, are an important delicacy and protein source in the diet of local Africans. In addition, plant infusions are used in traditional medicine to treat syphilis, dysentery, diarrhea, and inflamed eyes.3 C. mopane has a unique habitat in the dry low-lying areas with extreme climatic conditions, where it is exposed to shallow, poorly drained, often alkaline soils, thus increasing the chances of finding original metabolites that are indispensable for drug development.4 The phytochemistry of C. mopane is complex. Previous investigations revealed the presence of phenolic compounds in the heartwood5 and a few diterpenoids in the bark, seeds, husks, © XXXX American Chemical Society and American Society of Pharmacognosy



RESULTS AND DISCUSSION The metabolites 1−11 were purified from the seeds, leaves, and husks of C. mopane by using similar isolation methods. The three CH2Cl2 extracts were submitted to a preliminary fractionation by open column chromatography on silica gel. Fractions were combined on the basis of their TLC patterns and were further purified by semipreparative HSCCC, Received: August 16, 2015

A

DOI: 10.1021/acs.jnatprod.5b00729 J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products

Article

Figure 1. Ellipsoid plot of compound 10. There are four molecules present in the asymmetric unit that are not related by symmetry. For clarity reasons only one out of the four is shown here. All four molecules have the same stereochemistry, but show slightly different conformations in the peripheral substituents.

(Table 1) were supported by 2D NMR (1H−1H COSY, HSQC, and HMBC) experiments. In the 13C NMR spectrum (Table 1), 20 carbon signals were observed, which were assigned to four methyls, eight methylenes (including one oxygenated, δC 73.2), three methines (including one olefinic, δC 115.0), three quaternary carbons (δC 33.4, 42.6, and 171.6), one oxygenated tertiary carbon (δC 78.4), and one carbonyl carbon (δC 174.2). The HMBC correlations of H2-16 (δH 4.77) to C-13 (δC 171.6), C-14 (δC 115.0), and C-15 (δC 174.2) and of H-14 (δH 5.84) to C-13 (δC 171.6), C-15 (δC 174.2), and C16 (δC 73.2) suggested the presence of an α,β-unsaturated γlactone moiety. Hence, three of the indices of hydrogen deficiency could be accounted for. The two remaining indices were attributed to two ring systems. The 1H NMR data (Table 1) displayed singlets for two geminal methyl groups at δH 0.86 and 0.91, as well as one tertiary (δH 1.03) and one secondary methyl group at δH 1.06 (d, J = 7.7 Hz), diagnostic of a labdane-type diterpenoid. The HMBC correlations from H-8 (δH 1.87), H-11b (δH 1.70), H3-17 (δH 1.06), and H3-20 (δH 1.04) to an oxygenated tertiary carbon (δC 78.4) established the location of an OH group at C-9. Thus, the gross structure of 1 was established. The absolute configuration was assigned as (5S,8S,9R,10S) on the basis of the Flack parameter [−0.04(6)] obtained by low-temperature (123 K) Cu Kα radiation X-ray crystallography (Figure 2). Compound 1 is an isomer of viterotulin A that has been reported previously from Vitex rotundifolia.15 Compound 2, obtained as a white powder, gave a molecular formula of C20H34O4 according to an [M + Na]+ ion at m/z 361.2359 (calcd for C20H34NaO4, 361.2359) in its HRESIMS. The 1H and 13C NMR data of 2 (Table 1) were similar to those of dihydrogrindelic acid (5), with the difference being the presence of an oxygenated methine (δH 3.43, dd, J = 2.5 and 1.0 Hz, δC 76.4) in 2 vs a methylene group in 5. HMBC correlations from the gem-dimethyl group (δH 0.83 and 0.96; H3-19 and H3-18, respectively) to the oxygenated methine carbon (δC 76.4) helped to establish the 2D structure of 2. A severe overlapping of critical resonances in the 1H NMR spectrum of 2 precluded assignment of the relative configuration via NOESY data. The presence of the OH-3 group enabled the preparation of the methyl 3-p-bromobenzoate 2a (Table S4, Supporting Information). Its NOESY spectrum provided an improved signal dispersion that permitted assignment of the relative configuration (Figure 3). Selective irradiation of H3-20 enhanced H3-17, H3-19, and H2-11, thus

semipreparative HPLC, and chromatography on ODS and Sephadex LH-20 columns. The seed extract afforded compounds 1−3, 5, 7, 10, and 11. Compounds 4, 6, and 9 were isolated from the leaves, while the husks yielded compound 8. The structures of known compounds were determined by comparing their spectroscopic data with literature values. They were identified as moponeol A (4),8 also named mopaneol A, and rel-8S,13R-dihydrogrindelic acid (5).7 1H and 13C NMR chemical shifts of 5 were slightly different from reported values. The relative configuration of 5 was assigned as rel-8S,13R-dihydrogrindelic acid from the 1D and 2D NOESY (Figures S23−S25, Supporting Information). The same stereoisomer as 5 has previously been erroneously denominated as 8S,13S-dihydrogrindelic acid.7 Compounds 10 and 11 were identified as cis-clerodane diterpenoids previously reported from the aerial parts of Haploppapus paucidentatu12,13 and the stems of Aristolochia brasiliensis,14 respectively. Both compounds were reported without absolute configuration assignment and specific rotations. The absolute configuration of 10 was assigned by X-ray diffraction analysis as 5R,8S,9R,10S on the basis of the Flack parameter [0.02(5)] obtained by lowtemperature (123 K) Cu Kα radiation (Figure 1). Compound 11 had the same relative configuration and similar specific rotation ([α]25D −19, c 0.4, MeOH) to 10 ([α]25D −30, c 0.2, MeOH). Thus, the absolute configuration of 11 is proposed as 5R,8S,9R,10S. Compound 1, obtained as colorless crystals, showed a sodium adduct ion [M + Na]+ at m/z 343.2257 in its HRESIMS. Together with the 13C NMR data, this indicated a molecular formula of C20H32O3 (calcd for C20H32NaO3, 343.2244), which was consistent with five indices of hydrogen deficiency. Assignments of the 1H and 13C NMR data of 1 B

DOI: 10.1021/acs.jnatprod.5b00729 J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products

Article

Table 1. 1H and 13C NMR Spectroscopic Data for Compounds 1, 2, 3, and 6 (CDCl3, 600 MHz for 1H and 150 MHz for 13C NMR, δ in ppm) 1 δH (J in Hz)

position

2 δC, type

δH (J in Hz)

3

6

δC, type

δH (J in Hz)

δC, type 27.1, CH2

77.6, CH

36.8, 43.1, 16.5, 29.2,

C CH CH2 CH2

2.00, ma

38.5, 96.2, 41.5, 28.3,

CH C C CH2

2.00, ma

38.0, CH2

2.24−2.36, m

80.8, CH 46.6, CH2

5.81, br m

171.3, C 114.9, CH

4.73, 0.79, 0.91, 1.10, 0.64,

174.2, 73.2, 15.0, 21.3, 24.4, 16.3,

1

1.43−1.36, ma

32.6, CH2

1.94, ma 1.08, ma

26.2, CH2

2

1.61, m

18.2, CH2

1.94, ma

24.2, CH2

3

1.47, ma 1.37, ma

41.5, CH2

1.63, m 3.43, dd (2.5, 1.0)

76.4, CH

1.66, ma 1.20, ddd (12.2, 4.3, 2.0) 1.91, dddd (15.5, 15.0, 4.3, 2.7) 1.66, ma 4.62, t (2.7)

37.6, 40.9, 16.7, 29.4,

C CH CH2 CH2

1.77, ma 1.44, ma 1.98, ma

39.3, 94.0, 41.6, 28.7,

CH C C CH2

22.2, CH2

δH (J in Hz) 1.65, m 1.14, tdb (13.3, 3.6) 1.78, ma 1.49, ma 3.19, dd (11.5, 4.2)

δC, type 24.5, CH2

30.3, CH2

76.6, CH

1.15, tdb (13.5, 4.0) 4 5 6 7

8 9 10 11

12

13 14 15 16 17 18 (α) 19 20 21 22 a

1.45, ma 1.49−1.42, ma 1.84, m 1.48, ma 1.87, ma

1.97, ddd (13.8, 12.0, 5.0) 1.70, ddd (13.8, 12.7, 4.5) 2.58, dddd (16.5, 12.0, 4.5, 1.5) 2.35, dddd (16.5, 12.7, 5.0, 1.5)

33.4, 47.5, 17.3, 29.2,

C CH CH2 CH2

35.7, 78.4, 42.6, 29.6,

CH C C CH2

1.36, ma 1.88, ma

1.99, ma

1.47, ma 1.98, ma

1.88, ma 22.6, CH2

1.92, m

38.8, CH2

1.81, m

5.84, quint (1.5)

171.6, C 115.0, CH

4.77, 1.06, 0.91, 0.86, 1.03,

174.2, 73.2, 17.9, 33.8, 21.8, 17.1,

d (1.5) d (7.7) s s s

1.88, ma 1.33−1.45, m 1.99, ma

C CH2 CH3 CH3 CH3 CH3

2.67, d (13.3) 2.43, d (13.3) 1.39, 1.03, 0.96, 0.83, 0.95,

s d (7.8) s s s

5.52, br d (5.5) 1.88, br dt (17.6, 5.5) 1.77, ma 1.49, ma 2.05, br d (13.3) 1.62, dd (12.4, 5.3) 1.49, ma

41.8, 144.7, 118.4, 31.4,

C C CH CH2

33.2, 36.9, 39.0, 33.9,

CH C C CH2

22.3, CH2

1.79, m 80.7, C 46.8, CH2 173.2, 27.8, 18.1, 28.4, 21.9, 18.2,

C CH3 CH3 CH3 CH3 CH3

2.71, d (15.4) 2.48, d (15.4) 1.42, 1.08, 0.88, 0.90, 1.00,

s d (7.7) s s s

2.11, s

171.7, 27.5, 18.1, 28.0, 21.5, 18.3, 171.2, 21.2,

C CH3 CH3 CH3 CH3 CH3 C CH3

d (1.3) d (6.8) s s s

C CH2 CH3 CH3 CH3 CH3

Overlapping signals. bApparent multiplicity.

trans-junction of the decalin ring. Finally, the rel-13Rconfiguration of 2 was confirmed by NOESY cross-peaks between H2-14 and the ortho-protons of the benzoyl ring (Figure 3). The absolute configuration of 2 was established by the modified Mosher’s method.16 Esterification of the C-15 methyl ester of 2 with (S)- and (R)-MTPA-Cl afforded 3-O(R)-MTPA and 3-O-(S)-MTPA esters, respectively (Figure S15, Supporting Information). The Δδ values (δS − δR), calculated for the diastereomeric MTPA esters and expressed in hertz, are reported in Figure 4. The resonances of the C-1, C-2, C-14, and C-20 protons were shielded in the (R)-MTPA derivative, while the resonances of H3-18, H-5, and H2-6 were deshielded. Opposite trends were observed for the (S)-MTPA ester, suggesting the R configuration at C-3. The 3R absolute configuration assigned by the modified Mosher’s method was corroborated by electronic circular dichroism (ECD) data. The ECD spectrum of 2a (Figure 5) showed a positive Cotton effect at 245 nm due to the π → π* transition of the pbromobenzoate moiety, corresponding to a UV absorption maximum at 241 nm (Figure S49, Supporting Information). The experimental ECD spectrum of 2a matched the ECD curve

Figure 2. Ellipsoid plot of compound 1.

establishing their cofacial orientation, arbitrarily assigned as β (Figure S14, Supporting Information). On the other hand, the β-equatorial orientation of H-3 (δH 4.78) was inferred by its J values (dd, J(H-3/H-2eq) = 3.4 Hz and J(H-3/H-2ax) = 3.0 Hz) and corroborated by dipolar coupling with both H3-18 and H319. NOESY correlations between H-5 and H3-18 confirmed the C

DOI: 10.1021/acs.jnatprod.5b00729 J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products

Article

Figure 3. Lowest energy conformer of the methyl p-bromobenzoate 2a after DFT minimization in the gas phase at the B3LYP/6-31G* level. Key NOESY correlations (blue arrows) for compound 2a are shown.

Figure 4. Configurational assignment for alcohols 2 and 6 by the modified Mosher’s method. 1H NMR Δδ (δS − δR) shifts for the 3-OS- and 3-O-R-MTPA esters of compounds 2 and 6 are reported in hertz.

calculated for the (3R,5S,8S,9R,10S,13R)-stereoisomer (Figure 5). The reliability of the absolute configuration assignment at C-3 by ECD was further corroborated by the calculated ECD spectrum of the C-3 epimer of 2a, which showed Cotton effects that are not reconcilable with the experimental spectrum (Figure S51, Supporting Information). Compound 3, obtained as colorless crystals, showed a molecular formula of C22H36O5, according to an [M + Na]+ ion at m/z 403.2473 (calcd for C22H36NaO5, 403.2455) in its HRESIMS. The 1H and 13C NMR data of 3 were similar to those of 2 (Table 1). The only difference observed was that the C-3 hydroxy group in 2 is replaced by an acetoxy group in 3, as deduced from the HMBC correlation of H-3 (δH 4.62 t, J = 2.6 Hz) with an ester carbonyl (δC 171.2). Hydrolysis of the C-3 ester function afforded a derivative for which the 1H and 13C NMR data were identical to those of 2. Compound 3 was assigned the same absolute configuration as 2 on the basis of comparable specific rotation values ([α]25D −23, c 0.1, MeOH vs [α]25D −21, c 0.1, MeOH for 2). The molecular formula of compound 6 was determined as C20H30O3, based on a sodium adduct ion in the HRESIMS (m/ z 341.2102 [M + Na]+; calcd for C20H30NaO3, 341.2087). This implied six indices of hydrogen deficiency. Analogous to 1, inspection of the NMR data of 6 (Table 1) showed the

Figure 5. Experimental ECD spectra of compounds 2a and 6a and the calculated ECD spectrum for the 3R,5S,8S,9R,10S,13R stereoisomer of 2a in MeOH.

presence of a 4-ethylfuran-2(5H)one moiety. The remaining resonances comprised (i) three tertiary methyl groups at δH 0.64 (s, H3-20), 0.91 (s, H3-18), and 1.10 (s, H3-19); (ii) a sixproton spin system composed of two vicinal diastereotopic methylene groups (δH 1.14/1.65, H2-1 and δH 1.78/1.49, H2-2) that were linearly coupled to an angular methine (δH 2.05, br d, J = 13.3, H-10) and to an oxygenated methine (δH 3.19, dd, J = 11.5, 4.2 Hz, H-3); and (iii) a four-proton spin system composed of a methylene group (δH 1.77/1.88, H2-7) scalarly coupled to a tertiary methine (δH 1.49, m, H-8) and to an olefinic methine at δH 5.52 (d, J = 4.0 Hz, H-6). These spectroscopic features, along with the remaining three indices of hydrogen deficiency required by the molecular formula, D

DOI: 10.1021/acs.jnatprod.5b00729 J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products

Article

Figure 6. Lowest energy conformer of compound 6 after DFT minimization in the gas phase at the B3LYP/6-31G** level. Key NOESY correlations (blue arrows) are shown on the corresponding 2D structure.

suggested an isolabdane-type structure. 1H−13C HMBC correlations of the geminal dimethyl group H3-18 and H3-19 to the oxygenated methine (δC 76.6, C-3) and to an sp2 quaternary carbon at δC 144.7 (C-5) confirmed the C-3−C-6 connectivity. The relative configurations of the four stereocenters of 6 were deduced from NOESY correlations (Figure 6) and J-coupling analysis. The J-couplings of H-3 (δH 3.19, dd, J = 11.5, 4.2 Hz) and H-10 (δH 2.05, br d, J = 13.3 Hz) together with their NOESY correlations (H-3/H-2eq, H-3/H-1ax, H-3/ H3-19; H-10/H-1eq, H-10/H-2ax, H-10/H3-18) clearly indicated their axial disposition on opposite faces of the A ring and a chairlike conformation of this ring. The NOESY correlations of H-10/H2-12 and H3-17/H3-20 supported the α-orientation of the C-9 side chain and a β-cis-orientation of the C-17 and C-20 vicinal methyl groups. Stereoisomers of 6 have been synthesized previously from labdane diterpenoids isolated from Solidago species.17,18 The absolute configuration of compound 6 was inferred by the modified Mosher’s method, following the same procedure described for compound 2 (vide supra, Figure S33, Supporting Information). Similar to 2, the C1, C-2, C-10, and C-12 protons in 6 were shielded in the (R)MTPA ester, while resonances belonging to H3-19 and H2-6 were deshielded (Figure 4). The reverse was observed for the (S)-MTPA ester, suggesting an R configuration at C-3. This stereochemical assignment was confirmed unequivocally by the good match of the ECD spectra of the 3-p-bromobenzoate derivatives 2a and 6a (Figure 5). The positive split Cotton effect near 240 nm in the ECD spectrum of 6a resulted from exciton coupling comprising the π → π* electronic transition of the double bond and the p-bromobenzoate moiety (Figure 7).19 This confirmed the α-equatorial orientation of the C-3 ester functionality and hence the 3R absolute configuration. Thus, the absolute configuration of 6 was assigned as (3R,8S,9R,10S). Compound 7 was obtained as white crystals. The HRESIMS exhibited an [M + Na]+ ion at m/z 343.2260 (calcd for C20H32NaO3, 343.2244), suggesting a molecular formula of C20H32O3 and five indices of hydrogen deficiency. In addition to the resonances characteristic for a clerodane diterpenoid, the 1 H and 13C NMR spectra of 7 revealed the presence of a trisubstituted double bond (δH 5.65, br s, δC 115.4 C-14; δC 160.9, C-13) conjugated with a carboxylic group (δC 168.9, C15). These spectroscopic data showed close similarities to those of 3,4-epoxyclerodan-13E-en-15-oic acid reported from Detarium microcarpum fruits.20 However, slight differences in chemical shifts of the resonances belonging to the decalin

Figure 7. Exciton chirality of the homoallylic p-bromobenzoate of compound 6a.

system suggested that 7 is a diastereoisomer of the latter. Crystals suitable for X-ray diffraction analysis of 7 were obtained by crystallization from methanol-d4 (Figure 8). Hence,

Figure 8. Ellipsoid plot of compound 7. There are two molecules present in the asymmetric unit that are not related by symmetry. Both molecules have the same stereochemistry, but show slightly different conformations in the peripherical substituents. For clarity reasons only one out of the two is shown here.

the absolute configuration of 7 was assigned as (3R,4S,5R,8S,9R,10S) on the basis of the Flack parameter [0.04(6)] obtained by low-temperature (123 K) Cu Kα radiation X-ray diffraction analysis. The crystal structure of 7 showed that two molecules of 7 are linked via two hydrogen bonds involving the acidic proton of the hydroxycarbonyl group and the carbonyl oxygen. E

DOI: 10.1021/acs.jnatprod.5b00729 J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products

Article

Table 2. 1H and 13C NMR Spectroscopic Data for Compounds 7, 8, and 9 (Methanol-d4 for 7, CDCl3 for 8 and 9; 600 MHz for 1 H, and 150 MHz for 13C NMR, δ in ppm) 7 position

δC, type

1

1.60, m

17.0, CH2

2

2.10, m 1.79, ddd (15.7, 7.0, 3.3) 2.95, d (3.3)

23.1, CH2

3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 a

8

δH (mult, J in Hz)

2.06, 1.37, 1.46, 1.40, 1.49,

62.30, CH 62.33, C 35.1, C 35.4, CH2

ma ma ma ma ma

28.3, CH2 37.3, 38.8, 43.8, 36.8,

1.13, t (4.2) 1.57, ma 1.52, ma 1.98−2.08, m

5.65, br s 2.14, 0.87, 1.33, 1.10, 0.90,

d (1.1) d (6.7) s s s

CH C CH CH2

34.0, CH2 160.9, C 115.4, CH 168.9, C 17.7, CH3 14.9, CH3 19.6, CH3 27.6 CH3 18.7, CH3

δH (mult, J in Hz) 1.60, 1.43, 2.00, 1.83, 2.94,

ma m m ddd (15.7, 7.2, 3.5) d (3.5)

2.00, 1.32, 1.49, 1.36, 1.46,

ma ma m ma ma

9 δC, type 17.9, CH2 23.6, CH2 61.9, 62.1, 35.5, 35.0,

CH C C CH2

28.32, CH2

1.08, t (4.5) 1.65, dd (10.1, 6.8)

37.4, 39.0, 43.8, 36.2,

2.33−2.22, m

22.3, CH2

5.84, quint (1.6) 4.75, 0.84, 1.33, 1.11, 0.91,

d (1.6) d (6.8) s s s

CH C CH CH2

171.1, C 115.1, CH 174.1, C 73.1, CH2 16.0, CH3 20.7, CH3 28.27, CH3 19.9, CH3

δH (mult, J in Hz) 1.60, 1.51, 2.21, 2.08, 5.60,

m ma m m t (3.6)

26.6, CH2

1.75, dtb (12.6, 2.6) 1.33, m 1.42−1.47, ma 1.52−1.46, ma 1.35, 1.55, 1.43, 2.02, 1.94,

ma m ma tdb (13.0, 4.8) tdb (13.0, 4.8)

5.67, s 2.16, 0.81, 4.08, 1.02, 0.74,

d (1.0) d (6.0) m s s

δC, type 18.2, CH2

122.0, CH 147.8, C 37.7, C 36.29, CH2 27.2, CH2 36.32, CH 38.7, C 46.3, CH 36.3, CH2 34.9, CH2 163.9, 114.8, 171.6, 19.4, 15.9, 62.9, 21.3, 18.3,

C CH C CH3 CH3 CH2 CH3 CH3

Overlapping signals. bApparent multiplicity.

Compound 8 was obtained as a colorless oil. The HRESIMS data exhibited an [M + Na]+ ion at m/z 341.2101 (calcd for C20H30NaO3, 341.2087) that differed from that of 7 by 2 mass units. The 1H and 13C NMR resonances of 8 closely resembled those of 7 (Table 2), except for the replacement of the resonance of H3-16 [δH 2.14 (d, J = 1.1 Hz), δC 17.7] in 7 by a hydroxymethylene group [δH 4.75 (d, J = 1.6 Hz), δC = 73.1] in 8. Long-range HMBC correlations from the latter to the resonances at δC 115.1 (H-14), 171.1 (C-13), 174.1 (C-15), and 22.3 (H2-12) indicated the presence of an α,β-unsaturated γ-lactone moiety. Thus, the 2D structure of 8 was assigned as depicted. The cis-junction of the decalin ring was inferred by the J value of H-10 (δH 1.08, t, J = 4.5 Hz), indicative of dihedral angles of ca. 60° with H-1ax and H-1eq. NOESY correlations of H-10 and H-8 with H2-12 indicated their cofacial β-orientation (Figure 9) and established the relative configuration at C-8 and C-9. On the other hand, the NOESY associations, H-3/H3-18, H-3/H3-20, and H3-18/H3-20, and the J value (d, 3.5 Hz) of H-3 indicated the α-equatorial orientation of H-3 and, hence, a β-epoxide moiety (Figure 9). A natural clerodane with the same constitution and relative configuration as 8 has been reported from Ageratina saltillensis.21 However, the NMR data of 8 did not match with the reported data. A similar discrepancy was found in the synthesis of the (+)-enantiomer of 8 ([α]25D +11, MeOH).22 Hence, on the basis of the levorotatory specific rotation of 8 ([α]25D −5, MeOH), the stereostructure of 8 is assumed to be new, with its absolute configuration assigned as (3S,4R,5S,8R,9S,10R). Conspicuously, compounds 7 and 8, isolated from seeds and leaves, respectively, share enantiomeric decalin moieties.

Figure 9. Lowest energy conformer of compound 8 after DFT minimization in the gas phase at the B3LYP/6-31G** level. Key NOESY correlations (blue arrows) are shown on the 2D structure.

Compound 9 showed the same mass and the same planar structure as compound 10 and was a diastereoisomer of the latter [HRESIMS m/z 343.2258 (calcd for C20H32NaO3, 343.2244)]. Diagnostic NOEs permitted the identification of the cofacial orientation of the three methyl groups, Me-17 (δH 0.81), Me-19 (δH 1.02), and Me-20 (δH 0.74), as well as the antifacial orientation of H-10 (δH 1.35), arbitrarily assigned as β (Figure 10). The trans-junction of the decalin system arising from these assignments was further supported by the 13C NMR chemical shift of Me-19 (δC 21.3) that was shielded as compared to the shift in a cis-clerodane (Me-19 of 8 δC 28.27). 2 3 , 2 4 Compound 9 showed the same rel(5R,8R,9S,10R) relative configuration assigned to 16,18dihydroxykolavenic acid lactone, as confirmed by the excellent agreement of the 13C NMR data belonging to their decalin moieties.25 F

DOI: 10.1021/acs.jnatprod.5b00729 J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products



Article

EXPERIMENTAL SECTION

General Experimental Procedures. Optical rotations were measured with a PerkinElmer polarimeter (model 341) equipped with a sodium lamp (589 nm) and a 10 cm microcell. UV spectra were recorded in n-hexane on a Hewlett-Packard 8453 spectrophotometer. ECD spectra were measured in MeOH on a Chirascan CD spectrometer and were analyzed with Pro-Data V2.4 software. NMR spectra were recorded on a Bruker 600 AVANCE II operating at 600 (1H) and 150 MHz (13C) or on a Bruker AVANCE III 500 MHz spectrometer equipped with a 1 mm TXI microprobe (1H and 2D NMR) or a 5 mm BBO probe (13C NMR) (Bruker BioSpin) operating at 500 (1H) and 125 MHz (13C). 2D NMR experiments were performed using standard Bruker programs. All measurements were carried out at 291.15 K. Chemical shifts are reported as δ values (ppm) with the residual solvent signal (for 500 MHz NMR with microprobe) or TMS (for 600 MHz NMR) as internal reference, J in Hz. Standard pulse sequences from the Topspin 2.1 software package were used. For HRESIMS, a microTOF ESIMS system (Bruker Daltonics) was used. Mass calibration was performed with a solution of 0.1% formic acid in 2-propanol−water (1:1) containing 5 mM NaOH. Mass spectra were recorded in the range m/z 150−1500 in positive ion mode with the aid of microTOF control software 1.1 (Bruker Daltonics). HPLCPDA-MS analyses were performed with an Agilent 1100 system consisting of a degasser, quaternary pump, column oven, and PDA detector connected to a Gilson 215 injector and to an Esquire 3000 plus ion trap mass spectrometer (Bruker Daltonics). Data acquisition and processing were performed using HyStar 3.0 software (Bruker Daltonics). Materials for column chromatography were silica (200− 300 mesh, Merck), Sephadex LH-20 (40−70 μm; Amersham Pharmacia Biotech AB, Uppsala, Sweden), and YMC-Gel ODS-A (50 μm; YMC, Milford, MA, USA). HSGF254 silica gel TLC plates (Merck) were used for analytical TLC. A Spectrum model (Dynamic Extractions, Slough, UK) multilayer coil-planet J-type centrifuge was used for hydrodynamic high-speed countercurrent chromatography (HSCCC) according to the instrumental setup described in a previous report.32 The operation conditions of HSCCC were the same as a previous report.32 Semipreparative HPLC was performed with an Agilent 1100 series instrument equipped with a PDA detector. Data acquisition and processing were performed using HyStar 3.2 software (Bruker Daltonics). Solvents used for extraction and preparative separation were obtained from Merck Chemical Co. (South Africa) and were analytical reagent (AR) grade. HPLC-grade MeOH, MeCN (Scharlau Chemie S.A.), and H2O (obtained by an EASY-pure II from Barnstead water purification system, Dubuque) were used for HPLC separations. Dimethyl sulfoxide (DMSO) (Scharlau) was used for dissolving the samples. CDCl3 (100 atom % D, Armar Chemicals, and 99.8 atom % D, Sigma-Aldrich) and methanol-d4 (99.8 atom % D, Sigma-Aldrich) were used for NMR data collection. Plant Material. The seeds, husks, and leaves of C. mopane were collected in the Messina Magisterial District (Limpopo Province), South Africa, in March 2012 (P.C. & L. Zietsman 5241). A voucher specimen (NMB 26692) was deposited in the herbarium of the National Museum, Bloemfontein, South Africa (NMB). Extraction and Isolation. The air-dried seeds (277 g), husks (369 g), and leaves (709 g) of C. mopane were separately powdered and extracted with CH2Cl2 (3 × 1.5 L, 3 × 3.0 L, and 3 × 3.0 L, respectively; each 24 h) at room temperature. The solvents were evaporated under vacuum at 40 °C to yield 76.0 g of the seed extract, 15.4 g of the husk extract, and 37.1 g of the leaf extract. The CH2Cl2 extract of the seeds (63 g) of C. mopane was separated by open column chromatography (85 × 5 cm) on silica gel (230−400 mesh) with a step gradient [CH2Cl2−MeOH, 40:1 (4.1 L), 20:1 (0.84 L), 10:1 (0.88 L), 5:1 (0.6 L), 2.5:1 (0.7 L), 1:1 (0.4 L), and 0:1 (1 L)] at a flow rate of 10 mL/min. Fractions were combined based on TLC analysis [Fr. A (14.3 g), Fr. B (17.0 g), Fr. C (5.2 g), Fr. D (0.9 g), Fr. E (1.4 g), Fr. F (1.0 g), Fr. G (0.6 g)]. A portion of fraction B (500 mg) was purified by HSCCC [nheptane−EtOAc−MeOH−H2O (19:1:19:1); lower phase as mobile phase; flow rate 3 mL/min; sample was dissolved in a mixture of the

Figure 10. Lowest energy conformer of compound 9 after DFT minimization in the gas phase at the B3LYP/6-31G** level. Key NOESY correlations (blue arrows) are shown on the 2D structure.

Diterpenoids of different skeletal types have shown promising antimicrobial activities.26,27 Compound 11 showed good activity against Escherichia coli ATCC 8739 (MIC 15.6 μg/mL), Staphylococcus aureus ATCC 25923 (MIC 15.6 μg/ mL), and Enterococcus faecalis ATCC 29212 (MIC 31.3 μg/ mL), while the other diterpenoids showed MIC values mostly at ≥62.5 μg/mL. Isolated compounds with antimicrobial activities of 64−100 μg/mL are considered as having some clinical relevance.28 Hence the good activity found for compound 11 shows some promise. The crude leaf extract was also found to be the most active against Staph. aureus (125 μg/mL, Table S1, Supporting Information). Given that structurally related diterpenoids have shown positive GABAA receptor modulatory properties,29 compounds 1−11 were tested for enhancement of GABA-induced chloride currents in a Xenopus oocytes model, but were found to possess only negligible activity. Labdane and clerodane diterpenoids represent a large group of secondary metabolites that have shown interesting biological activities. Structurally, they have four or five contiguous stereocenters on the decalin system. According to biosynthetic arguments,30 natural labdane diterpenoids possess trans-fused decalin moieties, although both C-5/C-10 enantiomers (synand ent-forms) have been reported. On the other hand, clerodane-type diterpenoids can exhibit both trans- and cis-ring fusion.31 For both compound classes, the stereochemistry at C8 and C-9 can be either cis or trans, thereby allowing many diasteroisomeric combinations. Severely overlapping NMR signals and the lack of strong chromophores render the stereochemical assignment of these compounds a challenging task, as documented by numerous literature discrepancies. Within this work, we have employed different approaches to assign relative and/or absolute configuration to a small series of diterpenoids that may be helpful for future unambiguous identification of structurally related compounds. Microscale derivatization along with selective 1D experiments have been employed to overcome the problem of severe overlapping of NMR signals that is characteristic for these molecules. The matching of spectroscopic data with energyminimized structures confirmed the reliability of assignments of the relative configurations. For eight out of 11 compounds their absolute configuration could be assigned by X-ray crystallography, Mosher’s method, or chemical correlations. Finally, it has been shown that introduction of a strong chromophore like the p-bromobenzoyl group may facilitate the assignment of the absolute configuration by ECD for compounds lacking suitable chromophores. G

DOI: 10.1021/acs.jnatprod.5b00729 J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products

Article

Fraction G (2.8 g) was separated on a Sephadex LH-20 column (36 × 3 cm) with MeOH at a flow rate of 3 mL/min to give fractions G1− G3. Subfraction G2 (1.9 g) was separated on a Sephadex LH-20 column (36 × 3 cm) with MeOH−CHCl3 (1:1) at a flow rate of 3 mL/min to give fractions G2.1−G2.2. Fraction G2.2 (1.75 g) was separated by open column chromatography (71 × 2.3 cm) on silica gel (230−400 mesh) with an isocratic solvent system [CHCl3−EtOAc− AcOH, 9:1:0.1] at a flow rate of 3 mL/min to give fractions G2.2.1− G2.2.12. Purification of fraction G2.2.11 (1.75 g) by a Sephadex LH-20 column (20 × 2.5 cm) with EtOAc at a flow rate of 3 mL/min gave 4 (65 mg). Subfraction G2.2.5 (70 mg) was separated by open column chromatography (8 × 2.5 cm) on ODS with a step gradient [H2O− MeOH, 6:4 (0.1 L), 1:1 (0.6 L), 4:6 (0.8 L), 3:7 (1.1 L)] at a flow rate of 3 mL/min to give G2.2.5.1−G2.2.5.7. Purification of the G2.2.5.6 (20 mg) fraction by semipreparative HPLC [SunFire C18 column (5 μm, 10 × 150 mm, Waters)] with a gradient [H2O (A), MeOH (B); 5% → 70% B (0−18 min), 70% B (18−20 min), 70% → 100% B (20−23 min), and 100% B (23−25 min); flow rate 4 mL/min; sample concentration 66 mg/mL in DMSO; injection volume 50−100 μL] gave 9 (8.0 mg, tR 20.5 min). The CH2Cl2 extract of the husks of C. mopane (15.0 g) was separated by open column chromatography (55 × 3.5 cm) on silica gel (230−400 mesh) with an isocratic solvent system [CH2Cl2−MeOH, 40:1 (4.5 L), and washed with EtOAc (2 L)] at a flow rate of 10 mL/ min. Fractions were combined based on TLC analysis [Fr. A (1.9 g), Fr. B (4.3 g), Fr. C (2.0 g), Fr. D (2.5 g), Fr. E (0.5 g), Fr. F (0.9 g), Fr. G (0.9 g), Fr. H (0.5 g), and Fr. I (0.7 g)]. Fraction B (4.3 g) was separated by open column chromatography (57.5 × 3.5 cm) on silica gel (230−400 mesh) with an isocratic solvent system [n-hexane− EtOAc, 8:2 (7.7 L)] at a flow rate of 3 mL/min to give fractions B1− B14. The B12 fraction (680 mg) was further purified on Sephadex LH20 (EtOAc) and then silica gel (230−400 mesh, column i.d. 15 × 2.5 cm, n-hexane−EtOAc, 6:4) to give 8 (160 mg). (+)-(5S,8S,9R,10S)-9-Hydroxylabd-13-ene-15,16-olide (1): colorless, fine crystal (EtOAc); [α]25D +22 (c 0. 1, MeOH); UV λmax (nhexane) (log ε) 203 (2.65), 222 (sh) nm; 1H and 13C NMR data, Table 1; HRESIMS m/z 343.2257 [M + Na]+ (calcd for C20H32NaO3, 343.2244). (−)-(3R,5S,8S,9R,10S,13R)-3-Hydroxy-9,13-epoxylabdane-15-oic acid (2): white, amorphous powder; [α]25D −21 (c 0. 1, MeOH); UV λmax (n-hexane) (log ε) 206 (2.88) nm; 1H and 13C NMR data, Table 1; HRESIMS m/z 361.2359 [M + Na]+ (calcd for C20H34NaO4, 361.2349). (−)-(3R,5S,8S,9R,10S,13R)-3-Acetoxy-9,13-epoxylabdane-15-oic acid (3): white crystals [upper phase of n-heptane−EtOAc−MeOH− H2O (3:1:3:1)]; [α]25D −50 (c 0. 1, MeOH); UV λmax (n-hexane) (log ε) 203 (2.85), 222 (sh) nm; 1H and 13C NMR data, Table 1; HRESIMS m/z 403.2473 [M + Na]+ (calcd for C22H36NaO5, 403.2455). (+)-(3R,8S,9R,10S)-3-Hydroxylabda-5,13-dien-15,16-olide (6): white, amorphous powder; [α]25D +38 (c 0.6, MeOH); UV λmax (nhexane) (log ε) 213 (2.95) nm; 1H and 13C NMR data, Table 1; HRESIMS m/z 341.2102 [M + Na]+ (calcd for C20H30NaO3, 341.2087). (+)-(3R,4S,5R,8S,9R,10S)-3,4-Epoxyclerodane-13E-en-15-oic acid (7): white crystals (methanol-d4); [α]25D +57 (c 0.2, MeOH); UV λmax (n-hexane) (log ε) 221 (3.12) nm; 1H and 13C NMR data, Table 2; HRESIMS m/z 343.2260 [M + Na]+ (calcd for C20H32NaO3, 343.2244). (−)-(3S,4R,5S,8R,9S,10R)-3,4-Epoxyclerodane-13-en-15,16-olide (8): colorless oil; [α]25D −5 (c 0. 3, MeOH); UV λmax (n-hexane) (log ε) 206 (3.25) nm; 1H and 13C NMR data, Table 2; HRESIMS m/z 341.2101[M + Na]+ (calcd for C20H30NaO3, 341.2087). (+)-rel-(5R,8R,9S,10R)-18-Hydroxycleroda-3,13E-dien-15-oic acid (9): colorless gum; [α]25D +45 (c 0.6, MeOH); UV λmax (n-hexane) (log ε) 206 (2.38) nm; 1H and 13C NMR data, Table 2; HRESIMS m/ z 343.2258 [M + Na]+ (calcd for C20H32NaO3, 343.2244). X-ray Diffraction Analysis. The crystal data and absolute configurations of 1, 7, and 10 were determined using data collected

upper (2.0 mL) and the lower phases (2.0 mL) for injection] to give 5 (160 mg). Fraction C (5.2 g) was separated by open column chromatography (33 × 2.3 cm) on silica gel (230−400 mesh) with an isocratic solvent system [n-hexane−EtOAc, 3:7 (1 L), and washed with EtOAc− MeOH, 1:1 (2 L)]. A flow rate of 5 mL/min was used, and subfractions C1−C3 were obtained. Subfraction C2 (3.0 g) was separated by HSCCC [n-heptane− EtOAc−MeOH−H2O (3:1:3:1); lower phase as mobile phase; flow rate 3 mL/min; sample was dissolved in a mixture of the upper (5.0 mL) and the lower phases (5.0 mL) for injection] to give fractions C2.1−C2.9. Purification of 11 (6.0 mg, tR 22.0 min) from subfraction C2.7 (24 mg) was by semipreparative HPLC [SunFire C18 column (5 μm, 10 × 150 mm, Waters)] using a gradient [H2O (A), MeCN (B); 55 → 68% B (0−5 min), 68 B (5−10 min), 68 → 87% B (10−15 min), 87 B (15−20 min), 87 → 91% B (20−21 min), 91 B (21−23 min), 91 → 100% B (23−24 min), 100% B (24−26 min); flow rate 4 mL/min]. The sample concentration was 80 mg/mL in DMSO, and the injection volume was 50 μL. Subfraction C3 (904 mg) was separated by HSCCC [n-heptane−EtOAc−MeOH−H2O (2:1:2:1); upper phase as mobile phase; flow rate 3 mL/min; sample was dissolved in a mixture of the upper (2.0 mL) and the lower phases (2.0 mL) for injection] to give fractions C3.1−C3.10. Compound 1 (5.0 mg) was crystallized from an EtOAc solution of fraction C3.5 (24.7 mg). Subfraction C3.2 (153 mg) was separated by HSCCC [n-heptane− EtOAc−MeOH−H2O (3:1:3:1); lower phase as mobile phase; flow rate 3 mL/min; sample was dissolved in a mixture of the upper (2.0 mL) and the lower phases (2.0 mL) for injection] to give fractions C3.2.1−C3.2.3. Compound 3 (15.0 mg) was precipitated from a solution of fraction C3.2.2 (20.0 mg) in the upper phase of n-heptane−EtOAc− MeOH−H2O (3:1:3:1). Fractions D (820 mg) and E (1.3 g) were separated by HSCCC [nheptane−EtOAc−MeOH−H2O (5:2:5:2); lower phase as mobile phase; flow rate 3 mL/min; sample was dissolved in a mixture of the upper (5.0 mL) and the lower phases (5.0 mL) for injection] to give fractions D1−D9 and E1−E8, respectively. Compound 7 (150 mg) was crystallized from a MeOH solution of fraction D6 (155 mg). Compound 7 (150 mg) was obtained by precipitation in a MeOH solution of fraction D6 (155 mg). Recrystallization from methanol-d4 afforded white crystals suitable for X-ray analysis. Subfraction E4 (831 mg) was further separated by HSCCC [nheptane−EtOAc−MeOH−H2O (6:5:6:5); lower phase as mobile phase; flow rate 3 mL/min; sample was dissolved in a mixture of the upper (2.0 mL) and the lower phases (2.0 mL) for injection] to give fractions E4.1−E4.6. Compound 10 (172 mg) was crystallized from a MeOH solution of fraction E4.3 (180 mg). Fraction G (588 mg) was separated by HSCCC [n-heptane− EtOAc−MeOH−H2O (6:5:6:5); lower phase as mobile phase; flow rate 3 mL/min; sample was dissolved in a mixture of the upper (2.0 mL) and the lower phases (2.0 mL) for injection] to give fractions G1−G8. Subfraction G7 (200 mg) was separated by HSCCC [nheptane−EtOAc−MeOH−H2O (2:1:2:1); lower phase as mobile phase; flow rate 3 mL/min; sample was dissolved in a mixture of the upper (2.0 mL) and the lower phases (2.0 mL) for injection] to give fractions G7.1−E7.10. Compound 2 (51.0 mg) was precipitated from a MeOH solution of fraction G7.8 (55 mg). The CH2Cl2 extract of the leaves (36.0 g) of C. mopane was separated by open column chromatography (51 × 4 cm) on silica gel (230−400 mesh) with an isocratic solvent system [CH2Cl2−MeOH, 40:1 (4.8 L), and washed with EtOAc (2 L)] at a flow rate of 10 mL/ min. Fractions were combined based on TLC analysis [Fr. A (5.8 g), Fr. B (8.3 g), Fr. C (5.3 g), Fr. D (2.2 g), Fr. E (1.8 g), Fr. F (3.3 g), Fr. G (2.8 g), Fr. H (0.4 g), and Fr. I (3.1 g)]. Fraction E (1.8 g) was separated by open column chromatography (15 × 2.5 cm) on silica gel (230−400 mesh) with an isocratic solvent system [n-hexane−EtOAc, 7:3 (3 L), and washed with EtOAc (1 L)] at a flow rate of 5 mL/min to give fraction E1−E5. Fraction E4 (220 mg) was further purified on a Sephadex LH-20 (column i.d. 36 × 3 cm, MeOH−CHCl3, 1:1) and silica gel (230−400 mesh, column i.d. 15 × 2.5 cm, n-hexane−EtOAc, 6:4) to give 6 (100 mg). H

DOI: 10.1021/acs.jnatprod.5b00729 J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products

Article

SpecDis v1.61.40 ECD spectra were calculated from the spectra of individual conformers according to their contribution calculated by Boltzmann weighting. Derivatization of Compounds 2 and 6. C-15 Methyl Ester of Compound 2. To a solution of 2 (1.3 mg; 3.8 μmol) in n-hexane (0.2 mL) and anhydrous MeOH (20 μL) was added TMSCHN2 (20 μL; 10% in n-hexane, 0.6 mol/L) under an argon atmosphere. The pale yellow solution was allowed to stand in a sealed HPLC vial, at room temperature, for 1 h. The reaction was quenched by adding HOAc (10 μL) and dried under N2 flow (99% yield). General Procedure for the Synthesis of p-Bromobenzoyl Derivatives 2a and 6a. To stirred solutions of the methyl esters of 2 (1.5 mg; 4.2 μmol) and 6 (1.0 mg; 3.1 μmol) in CH2Cl2 (0.5 mL) were added p-bromobenzoyl chloride (2.5 equiv), Et3N (2.5 equiv), and 4-(dimethylamino)pyridine (0.5 equiv). The mixtures were stirred in sealed HPLC vials at room temperature for 12 h, and the reactions were quenched by adding 1 N HCl (0.5 mL). The organic phases were washed with NaHCO3 solution (0.5 mL) and brine (0.5 mL). After drying over Na2SO4 and filtration, the solvents were removed under N2. The p-bromobenzoyl derivatives 2a and 6a were purified by semipreparative HPLC [H2O (A), MeOH (B); 20% → 100% B (0−20 min, 20−25 min) and 100% B; flow rate 4 mL/min; detection at 254 nm] (75−80% yield). NMR data of 2a and 6a are provided in the Supporting Information (Table S4, Figures S10−S14, Figure S32, respectively). Compound 2a: ECD (MeOH, c 0.4 mM, 0.1 cm), Δε +1.9 (203 nm), −3.2 (218 nm), +1.6 (244). Compound 6a: ECD (MeOH, c 0.4 mM, 0.1 cm), Δε +13.5 (204 nm), −2.3 (219 nm), +2.1 (237). General Procedure for the Synthesis of Mosher Esters of 2 and 6. To solutions of the methyl esters of 2 (1.5 mg; 4.2 μmoL) and 6 (1.0 mg; 3.1 μmoL) in anhydrous CH2Cl2 (0.2 mL) were added (R)(−)- and (S)-(+)-MTPA chloride (2.0 equiv), anhydrous pyridine (10 equiv), and 4-(dimethylamino)pyridine (0.5 equiv). The mixtures were allowed to stand in sealed HPLC vials, at room temperature, for 16 h (or 80 h for compound 2). The reactions were then quenched by adding 1 N HCL (0.5 mL). Next, the organic phases were washed with NaHCO3 (0.5 mL) and brine (0.5 mL). After drying over Na2SO4 and filtration, the solvents were removed under N2. The Mosher esters of 2 and 6 were purified by semipreparative HPLC [H2O (A), MeOH (B); 40% → 100% B (0−20 min, 20−25 min) and 100% B; flow rate 4 mL/ min; detection at 254 nm] (60−70% yield). 1H and HSQC NMR spectra are provided as Supporting Information (Figures S15 and S33). Antimicrobial Activity. Extracts and compounds were evaluated for their antibacterial activity against four reference bacterial pathogens (Escherichia coli ATCC 8739, Klebsiella pneumoniae ATCC 13883, Staphylococcus aureus ATCC 25923, and Enterococcus faecalis ATCC 29212) (Table 3). The C. mopane crude extracts (starting concentration of 32 mg/mL) and isolated compounds (starting concentration of 5 mg/mL) were reconstituted in acetone and quantitatively evaluated for antimicrobial activity using the minimum inhibitory concentration (MIC) assay.41 Briefly, each well of a 96-well microtiter plate was filled with 100 μL of sterilized distilled H2O, and extracts and compounds (100 μL) were introduced into the wells of the first row. Serial doubling dilutions were performed. Cultures were added (100 μL) into all wells, yielding an inoculum size of approximately 1 × 106 colony forming units (CFU)/mL. Plates, sealed with a sterile adhesive sealer, were incubated at 37 °C for 24 h. After incubation, 40 μL of the color indicator, p-iodonitrotetrazolium violet (0.4 mg/mL; Sigma-Aldrich), was added to each well, which turned purple-pink in the presence of microbial growth. The end point MIC value was considered as the lowest concentration demonstrating no bacterial growth. The positive control (to confirm antimicrobial susceptibility) was ciprofloxacin (0.01 mg/mL). A negative control was included containing media and acetone to determine the effect of solvent on microbial growth. Finally, a culture control was included to confirm viability of microorganisms.

on a Bruker Kappa Apex2 diffractometer at 123 K using Cu Kα radiation with λ = 1.5418 Å, θmax = 68.127°, 70.075°, and 68.940°, for 1, 7, and 10, respectively. The Flack parameter33 was refined to comparable values in all three structures. Leverage analyses34 have been used to improve the accuracy of the obtained results, and standard uncertainties of the Flack parameter are in all three cases in the range of what can be expected for a measurement with Cu radiation from a light atom structure. The Apex2 software35 was used for data collection and integration. The structures were solved by the charge flipping method using the program Superflip.36 Least-squares refinement against F was carried out on all non-hydrogen atoms using the program CRYSTALS.37 Plots were produced using MERCURY.38 Crystal data of 1 (EtOAc): formula C20H32O3, M = 320.47, F(000) = 704, colorless block, size 0.050 × 0.130 × 0.190 mm3, orthorhombic, space group P212121, Z = 4, a = 7.6689(5) Å, b = 10.2286(6) Å, c = 22.6080(14) Å, α = 90°, β = 90°, γ = 90°, V = 1773.42(19) Å3, Dcalc = 1.200 Mg m−3. Minimal/maximal transmission 0.92/0.97, μ = 0.616 mm−1. From a total of 19 188 reflections, 3226 were independent (merging r = 0.024). From these, 3226 were considered as observed (I > 2.0σ(I)) and were used to refine 209 parameters, R = 0.0304 (observed data), wR = 0.0292 (all data), GOF = 0.9540. Minimal/ maximal residual electron density = −0.14/0.23 e Å−3. Flack parameter −0.04(6). Crystal data of 7(methanol-d4): formula C20H32O3, M = 320.47, F(000) = 704, colorless block, size 0.050 × 0.130 × 0.190 mm3, orthorhombic, space group P212121, Z = 4, a = 7.6689(5) Å, b = 10.2286(6) Å, c = 22.6080(14) Å, α = 90°, β = 90°, γ = 90°, V = 1773.42(19) Å3, Dcalc = 1.200 Mg m−3. Minimal/maximal transmission 0.92/0.97, μ = 0.616 mm−1. The Apex2 suite has been used for data collection and integration. From a total of 19 188 reflections, 3226 were independent (merging r = 0.024). From these, 3226 were considered as observed (I > 2.0σ(I)) and were used to refine 209 parameters, R = 0.0304 (observed data), wR = 0.0292 (all data), GOF = 0.9540. Minimal/maximal residual electron density = −0.14/0.23 e Å−3. Flack parameter 0.04(6). Crystal data of 10(MeOH): formula C20H32O3, M = 320.47, F(000) = 2816, colorless needle, size 0.050 × 0.110 × 0.240 mm3, orthorhombic, space group P212121, Z = 16, a = 14.0499(7) Å, b = 17.2599(9) Å, c = 31.2403(16) Å, α = 90°, β = 90°, γ = 90°, V = 7575.8(7) Å3, Dcalc = 1.124 Mg m−3. Minimal/maximal transmission 0.94/0.97, μ = 0.577 mm−1. From a total of 79 503 reflections, 13 943 were independent (merging r = 0.052). From these, 13 079 were considered as observed (I > 2.0σ(I)) and were used to refine 830 parameters, R = 0.0424 (observed data), wR = 0.0413 (all data), GOF = 0.9664. Minimal/maximal residual electron density = −0.14/0.19 e Å−3. Flack parameter 0.02(5). Crystallographic data for the structures have been deposited with the Cambridge Crystallographic Data Center; the deposition numbers are 1049108−1049110. The data can be obtained free of charge at http://www.ccdc.cam.ac.uk/. The crystal structure of 3 could not be elucidated. Numerous attempts were made to grow crystals and to measure X-ray diffraction data. It is likely that they were badly twinned, but in all cases the attempts to solve the structure were not successful. Computational Methods. Conformational analyses of compounds 2, 2a, 6, 6a, 8, and 9 were performed with Schrödinger MacroModel 9.8 (Schrödinger, LLC, New York) employing the OPLS2005 (optimized potential for liquid simulations) force field in H2O. Conformers within a 2 kcal/mol energy window from the global minimum were selected for geometrical optimization and energy calculation applying DFT with Becke’s nonlocal three-parameter exchange and correlation functional and the Lee−Yang−Parr correlation functional level (B3LYP) using the B3LYP/6-31G** or B3LYP/6-31G* (for 2a) basis set in the gas phase with the Gaussian 09 program package.39 Vibrational evaluation was done at the same level to confirm minima. For the 3-p-bromobenzoate derivatives 2a and 6a the excitation energy (denoted by wavelength in nm), rotatory strength dipole velocity (Rvel), and dipole length (Rlen) were calculated in MeOH by TD-DFT/B3LYP/6-31G**, using the SCRF method, with the CPCM model. ECD curves were obtained on the basis of rotatory strengths with a half-band of 0.16 eV and UV shift using I

DOI: 10.1021/acs.jnatprod.5b00729 J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products

Article

(2) Palgrave, K. C. Trees of Southern Africa; C. Struik: Cape Town, 1983; pp 268−269. (3) Watt, J. M.; Breyer-Brandwijl, M. G. The Medicinal Poisonous Plants of Southern and Eastern Africa; Livingstone: London, 1962. (4) Venter, F.; Venter, J. A. Making the Most Indigenous Trees; Briza Publications: Pretoria, 2002; p 100. (5) Ferreira, D.; Marais, J. P. J.; Slade, D. Phytochemistry 2003, 64, 31−51. (6) Mebe, P. P. Phytochemistry 2001, 57, 537−541. (7) Englund, B. M.; Griffeth, L.; Clausen, T. P. Phytochem. Lett. 2009, 2, 144−147. (8) Reiter, E.; Treadwell, E.; Cederstrom, E.; Reichardt, P. B.; Clausen, T. P. J. Nat. Prod. 2003, 66, 30−33. (9) Du, K.; Marston, A.; van Vuuren, S. F.; van Zyl, R. L.; Coleman, C.; Zietsman, P. C.; Bonnet, S. L.; Ferreira, D.; van der Westhuizen, J. H. Phytochem. Lett. 2014, 10, 123−128. (10) Hata, Y.; Raith, M.; Ebrahimi, S. N.; Zimmermann, S.; Mokoka, T.; Naidoo, D.; Fouche, G.; Maharaj, V.; Kaiser, M.; Brun, R.; Hamburger, M. Planta Med. 2013, 79, 492−498. (11) Hata, Y.; Ebrahimi, S. N.; De Mieri, M.; Zimmermann, S.; Mokoka, T.; Naidoo, D.; Fouche, G.; Maharaj, V.; Kaiser, M.; Brun, R.; Potterat, O.; Hamburger, M. Fitoterapia 2014, 93, 81−87. (12) Jakupovic, J.; Baruah, R. N.; Zdero, C.; Eid, F.; Pathak, V. P.; Chau-Thi, T. V.; Bohlmann, F.; King, R. M.; Robinson, H. Phytochemistry 1986, 25, 1873−1881. (13) Borquez, J.; Mancilla, A.; Pedreros, S.; Loyola, L. A.; Morales, G.; Wittke, O.; Brito, I. Bol. Soc. Chil. Quim. 1995, 40, 157−162. (14) Lopes, L. M. X.; Bolzani, V. D.; Trevisan, L. M. V. Phytochemistry 1987, 26, 2781−2784. (15) Lee, C.; Lee, J. W.; Jin, Q.; Lee, H. J.; Lee, S.-J.; Lee, D.; Lee, M. K.; Lee, C. K.; Hong, J. T.; Lee, M. K.; Hwang, B. Y. Bioorg. Med. Chem. Lett. 2013, 23, 6010−6014. (16) Hoye, T. R.; Jeffrey, C. S.; Shao, F. Nat. Protoc. 2007, 2, 2451− 2458. (17) Anderson, A. B.; McCrindle, R.; Nakamura, E. J. Chem. Soc., Chem. Commun. 1974, 453−454. (18) McCrindle, R.; Nakamura, E. Can. J. Chem. 1974, 52, 2029− 2036. (19) Manabe, S.; Enoki, N.; Nishino, C. Tetrahedron Lett. 1985, 26, 2213−2216. (20) Cavin, A. L.; Hay, A. E.; Marston, A.; Stoeckli-Evans, H.; Scopelliti, R.; Diallo, D.; Hostettmann, K. J. Nat. Prod. 2006, 69, 768− 773. (21) Fang, N.; Yu, S.; Mabry, T. J.; Abboud, K. A.; Simonsen, S. H. Phytochemistry 1988, 27, 3187−3196. (22) Piers, E.; Breau, M. L.; Han, Y.; Plourde, G. L.; Yeh, W.-L. J. Chem. Soc., Perkin Trans. 1 1995, 963−966. (23) Pacheco, A. G.; Machado de Oliveira, P.; Piló-Veloso, D.; Flávio de Carvalho Alcântara, A. Molecules 2009, 14, 1245−1262. (24) Manabe, S.; Nishino, C. Tetrahedron 1986, 42, 3461−3470. (25) Cao, S.; Hou, Y.; Brodie, P.; Miller, J. S.; Randrianaivo, R.; Rakotobe, E.; Rasamison, V. E.; Kingston, D. G. I. Chem. Biodiversity 2011, 8, 643−650. (26) Bisio, A.; Schito, A. M.; Ebrahimi, S. N.; Hamburger, M.; Mele, G.; Piatti, G.; Romussi, G.; Dal Piaz, F.; De Tommasi, N. Phytochemistry 2015, 110, 120−132. (27) García-Sánchez, E.; Ramírez-López, C. B.; Talavera-Alemán, A.; León-Hernández, A.; Martínez-Muñoz, R. E.; Martínez-Pacheco, M. M.; Gómez-Hurtado, M. A.; Cerda-García-Rojas, C. M.; JosephNathan, P.; del Río, R. E. J. Nat. Prod. 2014, 77, 1005−1012. (28) Gibbons, S. Nat. Prod. Rep. 2004, 21, 263−277. (29) Schramm, A.; Ebrahimi, S. N.; Raith, M.; Zaugg, J.; Rueda, D. C.; Hering, S.; Hamburger, M. Phytochemistry 2013, 96, 318−329. (30) Peters, R. J. Nat. Prod. Rep. 2010, 27, 1521−1530. (31) Tokoroyama, T. Synthesis 2000, 2000, 611−633. (32) Shikanga, E. A.; Viljoen, A.; Combrinck, S.; Marston, A. Phytochem. Lett. 2011, 4, 190−193. (33) Flack, H. Acta Crystallogr., Sect. A: Found. Crystallogr. 1983, 39, 876−881.

Table 3. Antibacterial Activities of Compounds 1−11 (MIC Values Are Given as μg/mL) compound 1 2 3 4 5 6 7 8 9 10 11 ciprofloxacin negative control culture control



K. pneumoniae ATCC 13883

E. coli ATCC 8739

E. faecalis ATCC 29212

Staph. aureus ATCC 25923

62.5 62.5 62.5 62.5 46.9 62.5 62.5 62.5 93.8 625 93.8 0.078 >8

250 125 125 125 125 125 >250 125 125 125 15.6 0.156 >8

125 62.5 62.5 125 125 62.5 62.5 62.5 125 62.5 31.3 0.156 >8

125 125 125 93.7 62.5 93.7 62.5 62.5 62.5 62.5 15.6 0.156 8

>8

>8

>8

>8

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jnatprod.5b00729. Antimicrobial activity of Colophospermum mopane crude extracts. 1H and 13C NMR shifts of known compounds 4, 5, 10, and 11. 1D and 2D NMR spectra of compounds 1−3, 6−9, 2a, and 6a. UV spectra of compounds 2a and 6a. Calculated ECD and UV spectra of the C-3 epimer of compound 2 (PDF) Crystallographic data (CIF) Crystallographic data (CIF) Crystallographic data (CIF)



AUTHOR INFORMATION

Corresponding Authors

*Tel: +41 61 267 14 25. Fax: +41 61 267 14 74. E-mail: [email protected]. *Tel: +27 828935113. Fax: +27-51-4448463. E-mail: [email protected]. Author Contributions ∞

K. Du and M. De Mieri contributed equally to this work.

Notes

The authors declare no competing financial interest. # Deceased on March 26, 2013.



ACKNOWLEDGMENTS ECD spectra were measured at the Biophysics Facility, Biozentrum, University of Basel. K.D. acknowledges the South African National Research Foundation (NRF) for financial support. We are grateful to Prof. S. Hering and Dr. S. Khom for the GABAA modulatory test. Thanks are due to Dr. S. Nejad Ebrahimi for helpful discussions.



REFERENCES

(1) Lewis, G.; Schrire, B.; Mackinder, B.; Lock, M. In Legumes of the World; Gardens, R. B., Eds.; E-Publishing Inc.: Kew, Richmond, Surrey, UK, 2005. J

DOI: 10.1021/acs.jnatprod.5b00729 J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products

Article

(34) Parsons, S.; Wagner, T.; Presly, O.; Wood, P. A.; Cooper, R. I. J. Appl. Crystallogr. 2012, 45, 417−429. (35) Apex2, Version 2, User Manual; Bruker Analytical X-ray Systems: Madison, WI, 2006. (36) Palatinus, L.; Chapuis, G. J. Appl. Crystallogr. 2007, 40, 786− 790. (37) Betteridge, P. W.; Carruthers, J. R.; Cooper, R. I.; Prout, K.; Watkin, D. J. J. Appl. Crystallogr. 2003, 36, 1487. (38) Macrae, C. F.; Bruno, I. J.; Chisholm, J. A.; Edgington, P. R.; McCabe, P.; Pidcock, E.; Rodriguez-Monge, L.; Taylor, R.; van de Streek, J.; Wood, P. A. J. Appl. Crystallogr. 2008, 41, 466−470. (39) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M. J.; Heyd, J.; Brothers, E. N.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A. P.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, N. J.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, Ö .; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09; Gaussian, Inc.: Wallingford, CT, USA, 2009. (40) Bruhn, T.; Hemberger, A. S. Y.; Bringmann, G. SpecDis version 1.61; University of Wuerzburg: Germany, 2014. (41) CLSI, Clinical and Laboratory Standards Institute. Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria that Grow Aerobically 22nd ed.; Approved Standard M100-S22, National Committee for Clinical Laboratory Standards: Fort Wayne, IN, 2012.

K

DOI: 10.1021/acs.jnatprod.5b00729 J. Nat. Prod. XXXX, XXX, XXX−XXX